Vulnerability
Ethnicity
The ethnic groups of all women who died were reported to the Enquiry, but the ethnic group of mothers in general is recorded only in England and not in the other countries of the United Kingdom. Since 1995, ethnic group information has been recorded in the Hospital Maternity Episode Statistics (HES) System for England, but coverage is still not complete. By the financial year 2004-05, ethnic group was recorded for 75% of deliveries in England for the years covered by this Report. A comparison of maternity HES data for 2000-01 with data about children under the age of one recorded in the 2001 census showed that the ethnic group distribution in HES delivery data was broadly comparable as long as maternities to women whose ethnic group was not stated are grouped with those to women whose ethnic group was recorded as White30. Maternity HES data for the financial years 2003-04 and 2004-05 have been grossed up to the total numbers of registered maternities in England in the 2003-05 triennium to produce the estimated maternities in Table 1.18. These have been used to produce estimated mortality rates and relative risks by ethnic group for England.
Table 1.18
Numbers and estimated rates of maternal death by type and ethnic group per 100,000 maternities;
England only: 2003-05.
| Ethnic group | Direct deaths | Indirect deaths | Total Direct and Indirect deaths | Estimated number of maternities |
||||||
|---|---|---|---|---|---|---|---|---|---|---|
| Numbers | Numbers | Numbers | Rate | 95 per cent CI for rate | Relative risk compared with white |
95 per cent CI for relative risk |
||||
| White | 71 | 91 | 162 | 11.1 | 9.5 | 12.9 | 1.0 | - | - | 1,462,537 |
| Mixed | 0 | 1 | 1 | 5.2 | 0.9 | 29.5 | 0.5 | 0.1 | 3.4 | 19,232 |
| Black African | 15 | 15 | 30 | 62.4 | 43.7 | 89.0 | 5.6 | 3.8 | 8.3 | 48,103 |
| Black Caribbean | 7 | 2 | 9 | 41.1 | 21.6 | 78.1 | 3.7 | 1.9 | 7.3 | 21,910 |
| Indian | 2 | 7 | 9 | 20.3 | 10.7 | 38.6 | 1.9 | 0.9 | 3.6 | 44,288 |
| Pakistani | 3 | 3 | 6 | 9.2 | 4.1 | 20.1 | 0.8 | 0.4 | 1.9 | 64,993 |
| Bangladeshi | 3 | 3 | 6 | 23.6 | 10.8 | 51.4 | 2.1 | 0.9 | 4.8 | 25,455 |
| Chinese and other Asian | 1 | 0 | 1 | 14.0 | 2.5 | 79.2 | 1.3 | 0.2 | 9.0 | 7,146 |
| Middle east | 4 | 3 | 7 | 32.0 | 15.5 | 66.1 | 2.9 | 1.4 | 6.2 | 21,845 |
| Other | 1 | 1 | 2 | 28.0 | 7.7 | 10.2 | 2.5 | 0.6 | 10.2 | 7,146 |
| Total | 107 | 126 | 233 | 13.5 | 11.9 | 15.4 | 1,722,655 | |||
These rates and relative risks are based on small numbers and the coding of ethnicity may be problematic so they should be interpreted with caution. Nevertheless, analysis of the English data suggests that for Black African women and, to a lesser extent Black Caribbean and Middle Eastern women, the mortality rate is significantly higher than that for White women. This may not only reflect the cultural factors implied in ethnicity but their social circumstances and the fact that some of them may have recently migrated to the United Kingdom under less than optimal circumstances. Table 1.19 shows the main causes of death by ethnic group, and Table 1.20 their access to care.
Table 1.19
Direct and Indirect maternal deaths by major ethnic group; England: 2003-05.
| White | Black African |
Black Caribbean |
Indian | Pakistani | Bangla- deshi |
Chinese and Asian |
Middle Eastern |
Other | Not stated |
Total |
|
|---|---|---|---|---|---|---|---|---|---|---|---|
| Direct | |||||||||||
| Thromboembolism | 30 | 3 | 3 | 0 | 2 | 0 | 1 | 1 | 1 | 0 | 41 |
| Pre-eclampsia / eclampsia | 12 | 2 | 2 | 0 | 0 | 1 | 1 | 1 | 0 | 0 | 19 |
| Haemorrhage | 7 | 2 | 1 | 1 | 2 | 1 | 1 | 1 | 1 | 0 | 17 |
| AFE | 11 | 4 | 0 | 1 | 0 | 1 | 0 | 0 | 0 | 0 | 17 |
| Early pregnancy | 12 | 2 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 14 |
| Sepsis | 14 | 3 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 18 |
| Anaesthetic | 5 | 0 | 0 | 0 | 0 | 0 | 1 | 0 | 0 | 0 | 6 |
| All Direct | 91 | 16 | 7 | 2 | 4 | 3 | 4 | 3 | 2 | 0 | 132 |
| Indirect | |||||||||||
| Cardiac | 33 | 6 | 1 | 4 | 0 | 1 | 2 | 0 | 1 | 0 | 48 |
| Other Indirect | 59 | 10 | 1 | 3 | 3 | 4 | 3 | 2 | 1 | 1 | 87 |
| Psychiatric Indirect | 16 | 2 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 18 |
| Indirect malignancies | 8 | 1 | 0 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 10 |
| All Indirect | 116 | 19 | 2 | 8 | 3 | 5 | 5 | 2 | 2 | 1 | 163 |
| Total | 207 | 35 | 9 | 10 | 7 | 8 | 9 | 5 | 4 | 1 | 295 |
Table 1.20
Characteristics of antenatal care sought by pregnant or recently delivered women by ethnic group and knowledge of English. Deaths from Direct and Indirect causes after 12 weeks or more weeks of gestation; United Kingdom: 2003-05.
| Ethnic group | Late or non attenders for antenatal care (ANC) | Total number of deaths after 12 or more weeks of gestation |
||
|---|---|---|---|---|
| Booked for care after 22 weeks or missed more than four visits |
No ANC | Subtotal | ||
| n | n | n(%) | n(%) | |
| White | 25 | 6 | 31 (17) | 183 (100) |
| Black African | 9 | 3 | 12 (40) | 30 (100) |
| Black Caribbean | 3 | 1 | 4 (57) | 7 (100) |
| Indian | 1 | 0 | 1 (11) | 9 (100) |
| Pakistani | 0 | 0 | 0 (0) | 5 (100) |
| Bangladeshi | 0 | 1 | 1 (13) | 8 (100) |
| Chinese and Asian | 1 | 0 | 1 (11) | 9 (100) |
| Middle Eastern | 1 | 0 | 1 (25) | 4 (100) |
| Other | 0 | 0 | 0 (0) | 3 (100) |
| Total | 40 | 11 | 51 (20) | 258 (100) |
| Women who did not speak English |
6 | 3 | 9 (35) | 26 (100) |
Newly arrived migrants, refugees and asylum seeking women
Women who have recently arrived into the UK, whatever their immigration status, bring new challenges for maternity services, some of which are seen in the UK for the first time in this Report. The key issues include poor overall health status, underlying and possible unrecognised medical conditions including congenital cardiac disease, and, increasingly HIV/AIDS and TB, all of which are classified as maternal deaths. Some suffered the consequences of genital mutilation. Others suffered the psychological and medical effects of fleeing war torn countries; four women raped by soldiers were too ashamed to admit to being pregnant or to seek maternity care on arrival in the UK. One other woman, who spoke no English, was kidnapped, raped and trafficked into the country to work as a prostitute then left on the street when her pregnancy became too advanced. Women who have been trafficked have fears about their status, language difficulties and do not know where to turn for help. Trafficked women also feel ashamed of being forced into sex work which transgresses their own cultural values and beliefs and make it difficult for them to reveal their situation31.
African migrant women
Of the 35 Black African women who suffered a maternal death, 16 died from Direct and 19 from Indirect causes, as shown in Table 1.21. Only four of the Black African women were UK citizens. Of the remaining 31, most were either recently arrived new immigrants, refugees or asylum seekers. In some cases their immigration status was not clear so these women are grouped together as “new arrivals”. This is more than double the number of cases in the last Report. There were an additional five women who appeared to have come to the UK to have NHS care for their pregnancy and childbirth. A further seven Black African women died within a year of childbirth, the majority from Late Indirect causes.
Apart from the five women who appeared to have come late in pregnancy expressly to have their baby in the UK, who came mainly from Nigeria, the African countries with the greatest representation of women were Somalia and Ethiopia, but a wide range of other countries were also represented including, for the first time, Francophone African countries.
Female Genital Cutting or Mutilation (FGC/M)
Although it is illegal to perform female circumcision, a procedure more commonly known as Female Genital Cutting (FGCiv) or Mutilation (FGC/M), in the UK it is likely that its prevalence amongst the pregnant population is increasing32. This is largely due to inward migration from countries or cultures where it is still routine practice, despite almost universal international condemnation at Government level. As well as being illegal to perform it in the UK, it is also illegal to perform it in any other country on UK citizens or permanent residents.
Estimates of prevalence in England and Wales have been derived by analysing births in England and Wales by the mother’s country of birth and applying estimates based on surveys to the prevalence of FGC/M in the country of origin. The estimated percentages of all maternities in England and Wales which were to women with FGC/M increased from 1.06 per cent in 2001 to 1.43 per cent in 2004. The latest study showed considerable geographical variation. In some major cities and other conurbations with large populations of Somali or Kenyan women, the estimated percentages of maternities affected by FGC/M had already exceeded two per cent by 200428.
FGC can affect women’s pregnancies in a number of ways33. This Report considered the deaths of at least four women known to have been cut in this way, three of whom did not disclose their condition until very late in pregnancy or in early labour. For one woman, her late disclosure may have directly contributed to her death following an unnecessary caesarean section because staff were not aware that a corrective procedure could have been performed during her antenatal period. For another, her condition was not apparent until she was first examined in established labour yet the obstetrician was not informed until her labour stalled some hours later. There are a growing number and variety of educational materials and sources of help available for both health professionals and women themselves.
In response to the increasing prevalence of women living with FGC/M in the UK, there are an increasing number of midwives and obstetricians and specialist services able to advise, help and support pregnant women living with FGC/M. As a result, women from countries where this is likely to be practiced should be sensitively asked about this during pregnancy and management plans agreed during the antenatal period.
Other migrant women
Although Black African women formed the majority of recently arrived women who died from maternal causes, there were also increases in, or lessons to be learnt from, other groups of new migrants.
Women who contract to marry UK men
Previous Reports have commented on how some women who contract to marry UK men in order to build a better life for themselves, who are sometimes demeaningly referred to as “mail order” brides, died because they were not helped by their husbands to seek the care they needed once pregnant. Language difficulties and cultural and geographical isolation meant they were also probably unaware of the maternity services available and where and how to access them. In this triennium at least one other such woman appears to have died for the same reasons. In previous Reports the majority of these women came from Thailand, but in this triennium three Vietnamese women also died.
New countries of the European Union (EU)
There were several maternal deaths of women who had recently arrived from countries newly admitted to the European Union. This reflects the experiences of the maternity services in general who report rising numbers of women from the expanded EU, many of whom do not speak English. As there are now 49 countries in Europe, with some of the best and poorest maternal health outcomes in the world, providing translation services is an ever increasing challenge. For the first time in this Report, there were a few maternal deaths in women from Turkey, an EU accession country.
“Health tourism”
Deaths amongst pregnant women who had specifically travelled to the UK for NHS health care is a new category for this Enquiry. There were at least five cases where women with pre-existing or past medical or obstetric complications came to the UK for NHS treatment. From the case details provided it is also possible six other women, who claimed to be refugees, may have entered the UK in order to obtain medical assistance. One European holiday maker, early in pregnancy, died of a pulmonary embolism shortly after her arrival.
Translation
Thirty-four women who died from maternal causes, another six who died from Coincidental causes, and eight who died some months after childbirth spoke little or no English. Very few had access to translation services and in most cases family members or friends were used as interpreters. Several of these were the woman’s own children, who may have been the only family members who could speak English. The use of family members or friends as translators causes concern because:
- The woman may be too embarassed to seek help for intimate concerns or discuss her past history
- It is not clear how much correct information was conveyed to the woman as the person who was interpreting did not have a good grasp of the language, or may have withheld information. In some cases the woman's pre-existing medical condition meant she was at considerable risk
- In some cases the translator was a perpetrator of domestic abuse against his partner, thus not enabling her to ask for advice or help
- It is not appropriate for a child to translate intimate details about his or her mother and unfair on both the woman and child.
Cultural practices and /or attitudes of male partners
Some women who died seemed to have been denied access to care because of cultural beliefs and practices where the responsibility for decision-making fell to their husbands or other family members. This was especially true for women who could not speak English. For example:
A pregnant woman who spoke no English had a port of arrival chest x-ray which was indicative of cardiac disease. Her English speaking husband was told by the port authorities that she was ill and he should take her to a doctor immediately. They had no means of communicating this directly to the woman who had, in any event, just landed in a strange airport in a distant country. Her husband did not tell her she was ill and she had no antenatal care at all. Her first contact with maternity services was when she was admitted in labour. She died of a post partum haemorrhage apparently still unaware of her possible cardiac disease.
For others, the actions of their partner may have been fatal:
Another non-English speaking woman had recurrent vaginal bleeding and was known to have a cervical fibroid. Heavy bleeding occurred in her second trimester and, although her husband said he would take her to hospital, he didn't. She collapsed at home the next day and required emergency admission. At this point she was unconscious, her wrists were found to have been grazed and bandaged, her haemoglobin was 7g and there was an intrauterine death. At hysterotomy a partially organised retroplacental clot was found. She had severe disseminated intravascular coagulation (DIC) and a CT scan showed contusions associated with small undisplaced fractures of her temporal and frontal bones. She died despite neurosurgical intervention for subdural haematoma.
In the autopsy report there was no mention of her wrist injuries, no reference to the DIC or its possible causes, and no examination of her skull. The report then identified a swollen brain with no signs of intracranial pressure.
There were instances when both partners and families were obstructive:
An extremely young, underage, bride who appears to have been brought into the country under false pretences became pregnant but had no antenatal care. She spoke no English and was not allowed out of her parents’ in-laws’ house. Her ‘husband’ eventually took her to see the GP in mid pregnancy, moribund from tuberculosis. Her in-laws lied about her age and showed no concern for her obvious ill-health or pregnancy and her “husband” told the GP he was not at all interested in obtaining antenatal care but wanted her treated because she was now too ill for sex. She died, in a caring hospital, alone in a strange country, only a few days later.
Socio-economic classification and employment
Since 2001, the National Statistics Socio-Economic Classification (NS SEC) has been used to classify social class in all official statistics in the United Kingdom. Because women’s occupations are missing from so many birth registration records, the NS SEC of the women’s husband or partner derived from his occupational code, is used in published tabulations of birth statistics for England and Wales. These are the data available for use as denominators for stillbirth, infant and maternal mortality rates. Therefore in order to calculate maternal mortality rates by NS SEC, the women’s husbands’ or partners’ occupations, where available, were used, irrespective of whether the woman’s own occupations were recorded. As numbers were small, the 3-class version of NS SEC was used. Where women were identified from their case notes as having no partner, this may have not been clear cut. For these women, the denominator used was sole registration by the mother alone, but it cannot be guaranteed that these women’s births would have been sole registrations had they survived.
Table 1.21
Numbers of maternal deaths by National Statistics Socio-Economic Classification (NS SEC) and rates per 100,000 estimated maternities; England and Wales: 2003-05.
| Social class of husband or partner and partnership status |
Direct | Indirect | Direct and indirect | Estimated maternities* | ||||||
|---|---|---|---|---|---|---|---|---|---|---|
| n | n | n (%) | Rate | 95 per cent CI for rate |
Relative risk | 95 per cent CI for relative risk |
||||
| Compared with managerial and professional | ||||||||||
| Managerial and professional | 27 | 25 | 52 (19) | 8.8 | 6.7 | 11.5 | 1.0 | 590,780 | ||
| Intermediate | 16 | 20 | 36 (13) | 9.0 | 6.5 | 12.4 | 1.0 | 0.7 | 1.6 | 401,520 |
| Routine and manual | 29 | 35 | 64 (23) | 9.7 | 7.6 | 12.4 | 1.1 | 0.8 | 1.6 | 659,310 |
| Compared with women with employed partners | ||||||||||
| All employed | 72 | 80 | 152 (55) | 9.2 | 7.8 | 10.8 | 1.0 | 1,651,610 | ||
| Unemployed, unclassifiable or not stated |
31 | 39 | 70 (26) | 68.5 | 54.2 | 86.6 | 7.4 | 5.6 | 9.9 | 102,150 |
| All women with partners | 103 | 119 | 222 (81) | 12.7 | 11.1 | 14.4 | 1,753,760 | |||
| Women without partners | 20 | 32 | 52 (19) | 38.6 | 29.4 | 50.6 | 4.2 | 3.1 | 5.7 | 134,743 |
| Employed | 11 | 4 | 15 (5) | |||||||
| Unemployed | 9 | 28 | 37 (14) | |||||||
| All women | 123 | 151 | 274 (100) | 14.5 | 12.9 | 16.3 | 1,888,487 | |||
For women with a husband or partner in employment, the social class differences in mortality were small and were no bigger than would be expected by chance, as shown in Table 1.21. In contrast, the rate for women whose partners were unemployed or whose occupations were unclassifiable was over seven times higher than that for all women with partners in employment. This group of women made up just over a quarter of the women in England and Wales who died, while only 55 per cent had partners in employment. Nearly a fifth of the women appeared to have no partner. Based on the tentative assumptions in the previous paragraph, their mortality rate was over four times that of women who had partners in employment.
Table 1.22 shows access to antenatal care by employment and partnership status.
Table 1.22
Characteristics of the antenatal care sought by pregnant or recently delivered women by employment and partnership. Direct and Indirect causes and gestation of 12 weeks of more; United Kingdom: 2003-05.
| Employment and relationship | Late or non attenders for antenatal care (ANC) | Number of deaths after 12 or more weeks of gestation |
||
|---|---|---|---|---|
| Booked after 22 weeks or missed more than four visits |
No ANC | Subtotal | ||
| n | n | n (%) | n (%) | |
| In partnership with at least one partner in employment |
8 | 1 | 9 (5) | 165 (100) |
| In partnership with neither partner in employment |
10 | 4 | 14 (47) | 30 (100) |
| Single woman in employment | 0 | 0 | 0 (0) | 14 (100) |
| Single woman unemployed | 16 | 3 | 19 (56) | 34 (100) |
| Total | 34 | 8 | 42 (17) | 243 (100) |
Area deprivation scores
As in the last Report, maternal deaths that occurred to residents of England have been analysed using the English Indices of Multiple Deprivation 2004. This is a score based on data for each small local area, known as a super output area, within electoral wards in England2. Postcodes are used to code addresses to super output areas and derive the scores. These are then ranked in to order and according to the score, and then grouped into quintiles. Of the 235 Direct and Indirect maternal deaths of women resident in England, 211 had a valid postcode at the time of death, and could therefore be included in the analysis in Table 1.23. Mortality rates and confidence intervals for each quintile are shown in Figure 1.3 and Table 1.23. This shows a clear gradient between the mortality rates for the least and most deprived areas and that mortality rates in the most deprived quintile were around five times higher than in the least deprived quintile. This was true for both Direct and Indirect causes of death independently.
Figure 1.3 Direct and Indirect maternal mortality rates and 95 per cent confidence intervals by deprivation quintile of place of residence; England: 2003-2005.
Table 1.23
Direct and Indirect deaths, rates per 100,000 maternities and relative risks by quintiles of place of residence of the Index of Multiple Deprivation 2004; England: 2003-05.
Quintile |
Number | Rate | 95 per cent CI for rate |
Relative risk | 95 per cent CI for relative risk |
Number of maternities |
||
|---|---|---|---|---|---|---|---|---|
| Direct | ||||||||
| Least deprived (1) | 10 | 2.0 | 1.1 | 3.8 | 1.0 | - | - | 490,102 |
| 2 | 17 | 4.5 | 2.8 | 7.2 | 2.2 | 1.0 | 4.8 | 378,896 |
| 3 | 20 | 6.2 | 4.0 | 9.5 | 3.0 | 1.4 | 6.4 | 325,057 |
| 4 | 24 | 8.0 | 5.4 | 11.9 | 3.9 | 1.9 | 8.2 | 301,043 |
| Most deprived (5) | 28 | 9.4 | 6.5 | 13.6 | 4.6 | 2.2 | 9.5 | 297,908 |
| All Direct | 99 | 5.5 | 4.5 | 6.7 | 1,793,006 | |||
| Indirect | ||||||||
| Least deprived (1) | 13 | 2.7 | 1.6 | 4.5 | 1.0 | - | - | 490,102 |
| 2 | 10 | 2.6 | 1.4 | 4.9 | 1.0 | 0.4 | 2.3 | 378,896 |
| 3 | 17 | 5.2 | 3.3 | 8.4 | 2.0 | 1.0 | 4.1 | 325,057 |
| 4 | 29 | 9.6 | 6.7 | 13.8 | 3.6 | 1.9 | 7.0 | 301,043 |
| Most deprived (5) | 43 | 14.4 | 10.7 | 19.4 | 5.4 | 2.9 | 10.1 | 297,908 |
| All Indirect | 112 | 6.2 | 5.2 | 7.5 | 1,793,006 | |||
| Direct and Indirect | ||||||||
| Least deprived (1) | 23 | 4.7 | 3.1 | 7.0 | 1.0 | - | - | 490,102 |
| 2 | 27 | 7.1 | 4.9 | 10.4 | 1.5 | 0.9 | 2.7 | 378,896 |
| 3 | 37 | 11.4 | 8.3 | 15.7 | 2.4 | 1.4 | 4.1 | 325,057 |
| 4 | 53 | 17.6 | 13.5 | 23.0 | 3.8 | 2.3 | 6.1 | 301,043 |
| Most deprived (5) | 71 | 23.8 | 18.9 | 30.1 | 5.1 | 3.2 | 8.1 | 297,908 |
| All | 211 | 11.8 | 10.3 | 13.5 | 1,793,006 | |||
Domestic abuse
Chapter 13 addresses the general issue of domestic abuse in pregnancy and how it affected the 70 women who were abused and whose deaths were considered by this Enquiry. For 19 women this abuse was fatal. The majority of the woman who suffered abuse during pregnancy had reported this to a maternity health professional. No woman appeared to have been routinely asked about abuse, a recent recommendation in these Reports, although the new national programme in England is only just being introduced. Eight women were living in a refuge having fled from abuse at home.
Even though the majority of these women died from causes unrelated to their pregnancy, there are lessons to be learnt from them concerning the identification and management of women living with abuse as well as the impact it is has on their ability to seek regular health care, as shown in Table 1.24. This shows that 81% of the women who died of Direct or Indirect causes, and who were in abusive relationships found it difficult to access or maintain contact with maternity services. Some of the others who were able to attend regularly had domineering partners who disrupted the relationship between the woman and her health care provider. Further, 77% of these women were in contact with their local social services and the child protection services were involved with 64% of the mothers and their children.
Table 1.24
Characteristics of the antenatal care sought by pregnant or recently delivered women who were murdered or known to be suffering domestic abuse; United Kingdom: 2003-05.
| Type of death | Death in early pregnancy |
Late or non attenders for antenatal care (ANC) | Total number of deaths of women* |
||
|---|---|---|---|---|---|
| Booked after 22 weeks or missed more than four visits |
No ANC | Subtotal | |||
| n | n | n | n (%) | n (%) | |
| Direct | 3 | 0 | 0 | 3 (75) | 4 (100) |
| Indirect | 1 | 6 | 3 | 10 (83) | 12 (100) |
| All | 4 | 6 | 3 | 13 (81) | 16 (100) |
| Coincidental | 2 | 4 | 3 | 9 (69) | 13 (100) |
| Late deaths | 0 | 15 | 2 | 17 (41) | 41 (100) |
| Total | 6 | 25 | 8 | 39 (56) | 70 (100) |
Child protection issues
Sixty-nine of the women whose deaths were reviewed this triennium were known to social services and /or child protection services (CPS), the vast majority of whom had previous children in care. Thirty women died from complications related to their pregnancy, i.e. from Direct or Indirect conditions, and half of the women who were murdered were also known by the CPS and/or social services.
As shown in Table 1.25, more than 80% of the women who died from Direct or Indirect did not seek care at all, booked late or failed to maintain regular contact with the maternity services, in the main because of fear that their unborn child might be removed at birth.
In forty-one cases a child protection case conference was held, 23 after the removal of the infant into care. Five women committed suicide shortly before or after their child’s case conference and another 18 women died from an apparent accidental overdose of a drug of abuse. Many of these women were late in booking and often defaulted from maternity care because of fear of social services involvement. In most cases, it appears that upon removal of the child the level of support and care from both maternity and social services fell or ceased. For example, in two such cases where an overdose of street drugs may have masked a suicide and a murder by a known abusive partner, the assessors said:
“Despite intensive support during pregnancy her care stopped in the post natal period. As soon as the baby was removed her care plans stopped, 24 hours after delivery.
“Once again it appears the care for the mother stops at the point at which the safety of the child is secured.”
Of all of the women who died and whose cases were assessed, 112 of their previous living children had already been adopted or placed in care. Several mothers had more than six previous children in care, the highest number being eight.
Table 1.25
Characteristics of the antenatal care sought by pregnant or recently delivered women who were known to the child protection services; United Kingdom: 2003-05.
| Type of death | Death in early pregnancy |
Late or non attenders for antenatal care (ANC) | Total number of deaths in women known to social services or the child protection services |
||
|---|---|---|---|---|---|
| Booked after 22 weeks or missed more than four visits |
No ANC | Subtotal | |||
| n | n (%) | n | n (%) | n (%) | |
| Direct | 3 | 3 | 2 | 8 (100) | 8 (100) |
| Indirect | 2 | 13 | 3 | 18 (75) | 24 (100) |
| All | 5 | 16 | 5 | 26 (81) | 32 (100) |
| Coincidental | 0 | 3 | 2 | 5 (45) | 11 (100) |
| Late Direct deaths | 0 | 1 | 0 | 1 (10) | 10 (100) |
| Total | 5 | 20 | 7 | 32 (60) | 53 (100) |
Substance misuse
Ninety-three of the women whose deaths were assessed this triennium had problems with substance misuse. Of these, 52 were drug addicts, another 32 women were occasional drug users and the remaining women were alcohol dependent. Seven died in very early pregnancy before they could access maternity care.
As seen in Table 1.26, as with domestic abuse and child protection, the majority of these women found it difficult to maintain contact with maternity services. However, for those who did, there was increasing evidence of a greater emphasis on planned multidisciplinary and multi-agency care, and in some cases the care they received was outstanding. Many of the women who received good or excellent care were in touch with their local Sure Start or Children’s Centre. Issues relating to substance misuse and pregnancy are discussed in Chapter 12.
Table 1.26
Characteristics of the antenatal care sought by pregnant or recently delivered women who were known substance misusers and whose pregnancy exceeded 12 weeks’ gestation; United Kingdom: 2003-05.
| Type of death | Late or non attenders for antenatal care (ANC) |
Total number of deaths of substance misusers after 12 or more weeks of gestation |
||
|---|---|---|---|---|
Booked after 22 weeks or missed more than four visits |
No ANC | Subtotal |
||
| n | n | n (%) | n (%) | |
| Direct | 3 | 2 | 5 (100) | 5 (100) |
| Indirect | 13 | 3 | 16 (73) | 22 (100) |
| All | 16 | 5 | 21 (78) | 27 (100) |
| Coincidental | 1 | 3 | 4 (33) | 12 (100) |
| Late deaths | 27 | 2 | 29 (62) | 47 (100) |
| Total | 44 | 10 | 54 (63) | 86 (100) |
Six women with substance abuse problems died early in pregnancy from Direct or Indirect causes, and there was one other Coincidental death. However, as shown in Table 1.26, 21 of the other 27 women whose pregnancies were advanced enough to require care, accessed maternity services less than optimally, or not at all. None of the women who died from Direct causes regularly attended care, and only 22% of those women who suffered an Indirect death attended for optimal antenatal care. By comparison, two thirds of the twelve women who died from Coincidental causes regularly attended for care.
Risk factors and barriers to care
Many of the women who died found it difficult to seek, or to maintain contact with, maternity and/or other health services. The many possible reasons for this have been discussed throughout this Chapter and the main characteristics of the women who found it difficult to attend are summarised in Table 1.27. Understanding what the barriers were that prevented these women from feeling able to access maternity care will help future services to develop in response to these needs. One of the key issues addressed in “Maternity Matters”, the modernisation agenda for maternity services in England3, is developing services that encourage and support all women, but particularly the most vulnerable, to access care early and stay in touch with it thereafter.
Table 1.27
Characteristics of the women who were poor or non attenders for antenatal care and whose pregnancy was 12 weeks of gestation or more; Direct and Indirect deaths; United Kingdom: 2003-05.
| Characteristic* | Women who were poor or non-attenders at antenatal care |
Overall number of women |
|---|---|---|
| n (%) | n (%) | |
| Domestic abuse | 13 (81) | 16 (100) |
| Known to child protection services or social services |
26 (81) | 32 (100) |
| Substance misuse | 21 (78) | 27 (100) |
| Black Caribbean | 4 (57) | 7 (100) |
| Single unemployed | 19 (56) | 34 (100) |
| Both partners unemployed | 14 (47) | 30 (100) |
| Black African | 12 (40) | 30 (100) |
| No English | 9 (35) | 26 (100) |
| White | 31 (17) | 183 (100) |
| At least one partner in employment | 9 (5) | 165 (100) |
What did you learn from this case and how has it changed your practice?
“I remember her lying in bed huddled up and crying and feeling frightened. In future I would like to believe that if I see this again I would drop everything and sit, listen and offer support.”
It is easy to forget that, apart from the partners, families and the communities of the women who died, every health care worker who knew or was involved in providing care for them was affected by their death. These rare events have a huge and long lasting impact on the staff involved. As part of participating in this Enquiry, every health care professional who was involved is asked “what did you learn from this case and how has it changed your practice?”. Many had never come across a maternal death before and all hope they would not do so again. A very few had to manage more than one, due entirely to the play of chance, and the impact of several deaths in a short period of time was immensely distressing for them.
Many thoughtful answers were provided, in general revealing the huge depth of caring and respect the staff had for the women who died, the babies and families who survived and the pain they themselves suffered as a result:
“The midwifery report made me cry.” (Central assessor)
“We all attended the funeral, even though she died some months after she left our care.” (A midwife who had cared for a terminally ill mother during her pregnancy).
Most staff reported learning clinical lessons which, in many cases, led to a change in personal or Trust based practice. Other staff were more reflective and philosophical:
“Even though she was a mother she was still a child herself.” (In relation to a young schoolgirl).
“How inspiring the human spirit can be in the face of insuperable odds.” (For a woman dying of cancer).
On the other hand a few professionals seemed oblivious to the poor quality of care they had provided, or to have had any understanding of the wider circumstances that may have affected a mother’s life and death. The assessors were also concerned by the apparently culturally dismissive or insensitive remarks made by a few professionals during the course of the reviews.
In some cases staff appeared to take the blame upon themselves despite providing the best possible care. In these cases the Enquiry assessors were saddened as they considered the care these professionals had provided was exemplary and that the failings in the system were totally outside the control of these workers. Some staff who cared for women who declined help, or who had to watch a woman bleed to death while refusing blood products, reported significant personal distress. It is important that all staff affected by a maternal death should be offered support and counselling to help them come to terms with their own reactions.